自我RAG

自我RAG是一种包含对检索文档和生成内容进行自我反思/自我评分的RAG策略。

在论文中，做出了以下决定：

1. 我是否应该从检索器 R 中检索 -

2. 输入：x (问题) 或 x (问题), y (生成)

3. 决定何时使用 R 检索 D 个片段

4. 输出：yes, no, continue

5. 检索到的段落 D 是否与问题 x 相关 -

6. 输入：(x (问题), d (片段)) 对于 D 中的每一个 d

7. d 提供了对解决 x 问题有用的资料

8. 输出：relevant, irrelevant

9. 大语言模型从每个 D 中的片段生成的内容是否与该片段相关（幻觉等） -

10. 输入：x (问题), d (片段), y (生成) 对于 D 中的每一个 d

11. y (生成) 中所有值得验证的陈述都得到了 d 的支持

12. 输出：{fully supported, partially supported, no support}

13. 大语言模型从每个 D 中的片段生成的内容是否是对 x (问题) 的有用响应 -

14. 输入：x (问题), y (生成) 对于 D 中的每一个 d

15. y (生成) 是对 x (问题) 的有用响应
16. 输出：{5, 4, 3, 2, 1}

我们将使用LangGraph从头开始实现其中的一些想法。

设置

首先，让我们安装所需的包并设置API密钥

%pip install -U langchain_community tiktoken langchain-openai langchainhub chromadb langchain langgraph

import getpass
import os


def _set_env(key: str):
    if key not in os.environ:
        os.environ[key] = getpass.getpass(f"{key}:")


_set_env("OPENAI_API_KEY")

为LangGraph开发设置LangSmith

注册LangSmith，可以快速发现并解决您的LangGraph项目中的问题，提高项目性能。LangSmith允许您使用跟踪数据来调试、测试和监控使用LangGraph构建的LLM应用程序——更多关于如何开始的内容，请参阅此处。

检索器

让我们索引3篇博客文章。

from langchain.text_splitter import RecursiveCharacterTextSplitter
from langchain_community.document_loaders import WebBaseLoader
from langchain_community.vectorstores import Chroma
from langchain_openai import OpenAIEmbeddings

urls = [
    "https://lilianweng.github.io/posts/2023-06-23-agent/",
    "https://lilianweng.github.io/posts/2023-03-15-prompt-engineering/",
    "https://lilianweng.github.io/posts/2023-10-25-adv-attack-llm/",
]

docs = [WebBaseLoader(url).load() for url in urls]
docs_list = [item for sublist in docs for item in sublist]

text_splitter = RecursiveCharacterTextSplitter.from_tiktoken_encoder(
    chunk_size=250, chunk_overlap=0
)
doc_splits = text_splitter.split_documents(docs_list)

# Add to vectorDB
vectorstore = Chroma.from_documents(
    documents=doc_splits,
    collection_name="rag-chroma",
    embedding=OpenAIEmbeddings(),
)
retriever = vectorstore.as_retriever()

API Reference: RecursiveCharacterTextSplitter | WebBaseLoader | Chroma | OpenAIEmbeddings

大型语言模型(LLMs)

使用Pydantic与LangChain

本笔记本使用Pydantic v2 BaseModel，这需要langchain-core >= 0.3。使用langchain-core < 0.3将会由于混用了Pydantic v1和v2的BaseModel而导致错误。

### Retrieval Grader


from langchain_core.prompts import ChatPromptTemplate
from langchain_openai import ChatOpenAI

from pydantic import BaseModel, Field


# Data model
class GradeDocuments(BaseModel):
    """Binary score for relevance check on retrieved documents."""

    binary_score: str = Field(
        description="Documents are relevant to the question, 'yes' or 'no'"
    )


# LLM with function call
llm = ChatOpenAI(model="gpt-4o-mini", temperature=0)
structured_llm_grader = llm.with_structured_output(GradeDocuments)

# Prompt
system = """You are a grader assessing relevance of a retrieved document to a user question. \n 
    It does not need to be a stringent test. The goal is to filter out erroneous retrievals. \n
    If the document contains keyword(s) or semantic meaning related to the user question, grade it as relevant. \n
    Give a binary score 'yes' or 'no' score to indicate whether the document is relevant to the question."""
grade_prompt = ChatPromptTemplate.from_messages(
    [
        ("system", system),
        ("human", "Retrieved document: \n\n {document} \n\n User question: {question}"),
    ]
)

retrieval_grader = grade_prompt | structured_llm_grader
question = "agent memory"
docs = retriever.invoke(question)
doc_txt = docs[1].page_content
print(retrieval_grader.invoke({"question": question, "document": doc_txt}))

API Reference: ChatPromptTemplate | ChatOpenAI

binary_score='no'

### Generate

from langchain import hub
from langchain_core.output_parsers import StrOutputParser

# Prompt
prompt = hub.pull("rlm/rag-prompt")

# LLM
llm = ChatOpenAI(model_name="gpt-4o-mini", temperature=0)


# Post-processing
def format_docs(docs):
    return "\n\n".join(doc.page_content for doc in docs)


# Chain
rag_chain = prompt | llm | StrOutputParser()

# Run
generation = rag_chain.invoke({"context": docs, "question": question})
print(generation)

API Reference: StrOutputParser

The design of generative agents combines LLM with memory, planning, and reflection mechanisms to enable agents to behave conditioned on past experience. Memory stream is a long-term memory module that records a comprehensive list of agents' experience in natural language. LLM functions as the agent's brain in an autonomous agent system.

### Hallucination Grader


# Data model
class GradeHallucinations(BaseModel):
    """Binary score for hallucination present in generation answer."""

    binary_score: str = Field(
        description="Answer is grounded in the facts, 'yes' or 'no'"
    )


# LLM with function call
llm = ChatOpenAI(model="gpt-4o-mini", temperature=0)
structured_llm_grader = llm.with_structured_output(GradeHallucinations)

# Prompt
system = """You are a grader assessing whether an LLM generation is grounded in / supported by a set of retrieved facts. \n 
     Give a binary score 'yes' or 'no'. 'Yes' means that the answer is grounded in / supported by the set of facts."""
hallucination_prompt = ChatPromptTemplate.from_messages(
    [
        ("system", system),
        ("human", "Set of facts: \n\n {documents} \n\n LLM generation: {generation}"),
    ]
)

hallucination_grader = hallucination_prompt | structured_llm_grader
hallucination_grader.invoke({"documents": docs, "generation": generation})

GradeHallucinations(binary_score='yes')

### Answer Grader


# Data model
class GradeAnswer(BaseModel):
    """Binary score to assess answer addresses question."""

    binary_score: str = Field(
        description="Answer addresses the question, 'yes' or 'no'"
    )


# LLM with function call
llm = ChatOpenAI(model="gpt-4o-mini", temperature=0)
structured_llm_grader = llm.with_structured_output(GradeAnswer)

# Prompt
system = """You are a grader assessing whether an answer addresses / resolves a question \n 
     Give a binary score 'yes' or 'no'. Yes' means that the answer resolves the question."""
answer_prompt = ChatPromptTemplate.from_messages(
    [
        ("system", system),
        ("human", "User question: \n\n {question} \n\n LLM generation: {generation}"),
    ]
)

answer_grader = answer_prompt | structured_llm_grader
answer_grader.invoke({"question": question, "generation": generation})

GradeAnswer(binary_score='yes')

### Question Re-writer

# LLM
llm = ChatOpenAI(model="gpt-4o-mini", temperature=0)

# Prompt
system = """You a question re-writer that converts an input question to a better version that is optimized \n 
     for vectorstore retrieval. Look at the input and try to reason about the underlying semantic intent / meaning."""
re_write_prompt = ChatPromptTemplate.from_messages(
    [
        ("system", system),
        (
            "human",
            "Here is the initial question: \n\n {question} \n Formulate an improved question.",
        ),
    ]
)

question_rewriter = re_write_prompt | llm | StrOutputParser()
question_rewriter.invoke({"question": question})

"What is the role of memory in an agent's functioning?"

图

以图的形式捕捉流程。

图状态

from typing import List

from typing_extensions import TypedDict


class GraphState(TypedDict):
    """
    Represents the state of our graph.

    Attributes:
        question: question
        generation: LLM generation
        documents: list of documents
    """

    question: str
    generation: str
    documents: List[str]

### Nodes


def retrieve(state):
    """
    Retrieve documents

    Args:
        state (dict): The current graph state

    Returns:
        state (dict): New key added to state, documents, that contains retrieved documents
    """
    print("---RETRIEVE---")
    question = state["question"]

    # Retrieval
    documents = retriever.invoke(question)
    return {"documents": documents, "question": question}


def generate(state):
    """
    Generate answer

    Args:
        state (dict): The current graph state

    Returns:
        state (dict): New key added to state, generation, that contains LLM generation
    """
    print("---GENERATE---")
    question = state["question"]
    documents = state["documents"]

    # RAG generation
    generation = rag_chain.invoke({"context": documents, "question": question})
    return {"documents": documents, "question": question, "generation": generation}


def grade_documents(state):
    """
    Determines whether the retrieved documents are relevant to the question.

    Args:
        state (dict): The current graph state

    Returns:
        state (dict): Updates documents key with only filtered relevant documents
    """

    print("---CHECK DOCUMENT RELEVANCE TO QUESTION---")
    question = state["question"]
    documents = state["documents"]

    # Score each doc
    filtered_docs = []
    for d in documents:
        score = retrieval_grader.invoke(
            {"question": question, "document": d.page_content}
        )
        grade = score.binary_score
        if grade == "yes":
            print("---GRADE: DOCUMENT RELEVANT---")
            filtered_docs.append(d)
        else:
            print("---GRADE: DOCUMENT NOT RELEVANT---")
            continue
    return {"documents": filtered_docs, "question": question}


def transform_query(state):
    """
    Transform the query to produce a better question.

    Args:
        state (dict): The current graph state

    Returns:
        state (dict): Updates question key with a re-phrased question
    """

    print("---TRANSFORM QUERY---")
    question = state["question"]
    documents = state["documents"]

    # Re-write question
    better_question = question_rewriter.invoke({"question": question})
    return {"documents": documents, "question": better_question}


### Edges


def decide_to_generate(state):
    """
    Determines whether to generate an answer, or re-generate a question.

    Args:
        state (dict): The current graph state

    Returns:
        str: Binary decision for next node to call
    """

    print("---ASSESS GRADED DOCUMENTS---")
    state["question"]
    filtered_documents = state["documents"]

    if not filtered_documents:
        # All documents have been filtered check_relevance
        # We will re-generate a new query
        print(
            "---DECISION: ALL DOCUMENTS ARE NOT RELEVANT TO QUESTION, TRANSFORM QUERY---"
        )
        return "transform_query"
    else:
        # We have relevant documents, so generate answer
        print("---DECISION: GENERATE---")
        return "generate"


def grade_generation_v_documents_and_question(state):
    """
    Determines whether the generation is grounded in the document and answers question.

    Args:
        state (dict): The current graph state

    Returns:
        str: Decision for next node to call
    """

    print("---CHECK HALLUCINATIONS---")
    question = state["question"]
    documents = state["documents"]
    generation = state["generation"]

    score = hallucination_grader.invoke(
        {"documents": documents, "generation": generation}
    )
    grade = score.binary_score

    # Check hallucination
    if grade == "yes":
        print("---DECISION: GENERATION IS GROUNDED IN DOCUMENTS---")
        # Check question-answering
        print("---GRADE GENERATION vs QUESTION---")
        score = answer_grader.invoke({"question": question, "generation": generation})
        grade = score.binary_score
        if grade == "yes":
            print("---DECISION: GENERATION ADDRESSES QUESTION---")
            return "useful"
        else:
            print("---DECISION: GENERATION DOES NOT ADDRESS QUESTION---")
            return "not useful"
    else:
        pprint("---DECISION: GENERATION IS NOT GROUNDED IN DOCUMENTS, RE-TRY---")
        return "not supported"

构建图

这仅仅是遵循了我们在上面的图中概述的流程。

from langgraph.graph import END, StateGraph, START

workflow = StateGraph(GraphState)

# Define the nodes
workflow.add_node("retrieve", retrieve)  # retrieve
workflow.add_node("grade_documents", grade_documents)  # grade documents
workflow.add_node("generate", generate)  # generatae
workflow.add_node("transform_query", transform_query)  # transform_query

# Build graph
workflow.add_edge(START, "retrieve")
workflow.add_edge("retrieve", "grade_documents")
workflow.add_conditional_edges(
    "grade_documents",
    decide_to_generate,
    {
        "transform_query": "transform_query",
        "generate": "generate",
    },
)
workflow.add_edge("transform_query", "retrieve")
workflow.add_conditional_edges(
    "generate",
    grade_generation_v_documents_and_question,
    {
        "not supported": "generate",
        "useful": END,
        "not useful": "transform_query",
    },
)

# Compile
app = workflow.compile()

API Reference: END | StateGraph | START

from pprint import pprint

# Run
inputs = {"question": "Explain how the different types of agent memory work?"}
for output in app.stream(inputs):
    for key, value in output.items():
        # Node
        pprint(f"Node '{key}':")
        # Optional: print full state at each node
        # pprint.pprint(value["keys"], indent=2, width=80, depth=None)
    pprint("\n---\n")

# Final generation
pprint(value["generation"])

---RETRIEVE---
"Node 'retrieve':"
'\n---\n'
---CHECK DOCUMENT RELEVANCE TO QUESTION---
---GRADE: DOCUMENT NOT RELEVANT---
---GRADE: DOCUMENT RELEVANT---
---GRADE: DOCUMENT NOT RELEVANT---
---GRADE: DOCUMENT RELEVANT---
---ASSESS GRADED DOCUMENTS---
---DECISION: GENERATE---
"Node 'grade_documents':"
'\n---\n'
---GENERATE---
---CHECK HALLUCINATIONS---
---DECISION: GENERATION IS GROUNDED IN DOCUMENTS---
---GRADE GENERATION vs QUESTION---
---DECISION: GENERATION ADDRESSES QUESTION---
"Node 'generate':"
'\n---\n'
('Short-term memory is used for in-context learning in agents, allowing them '
 'to learn quickly. Long-term memory enables agents to retain and recall vast '
 'amounts of information over extended periods. Agents can also utilize '
 'external tools like APIs to access additional information beyond what is '
 'stored in their memory.')

inputs = {"question": "Explain how chain of thought prompting works?"}
for output in app.stream(inputs):
    for key, value in output.items():
        # Node
        pprint(f"Node '{key}':")
        # Optional: print full state at each node
        # pprint.pprint(value["keys"], indent=2, width=80, depth=None)
    pprint("\n---\n")

# Final generation
pprint(value["generation"])

---RETRIEVE---
"Node 'retrieve':"
'\n---\n'
---CHECK DOCUMENT RELEVANCE TO QUESTION---
---GRADE: DOCUMENT RELEVANT---
---GRADE: DOCUMENT NOT RELEVANT---
---GRADE: DOCUMENT RELEVANT---
---GRADE: DOCUMENT RELEVANT---
---ASSESS GRADED DOCUMENTS---
---DECISION: GENERATE---
"Node 'grade_documents':"
'\n---\n'
---GENERATE---
---CHECK HALLUCINATIONS---
---DECISION: GENERATION IS GROUNDED IN DOCUMENTS---
---GRADE GENERATION vs QUESTION---
---DECISION: GENERATION ADDRESSES QUESTION---
"Node 'generate':"
'\n---\n'
('Chain of thought prompting works by repeatedly prompting the model to ask '
 'follow-up questions to construct the thought process iteratively. This '
 'method can be combined with queries to search for relevant entities and '
 'content to add back into the context. It extends the thought process by '
 'exploring multiple reasoning possibilities at each step, creating a tree '
 'structure of thoughts.')

Comments